Transporting Lung Cancer Screening Effects From National Lung Screening Trial to a US Screening-Eligible Population
7 affiliations
- 1Department of Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill, NC
- 2Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC
- 3Department of Radiology, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC
- 4Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC
- 5Division of General Medicine and Clinical Epidemiology, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC
- 6Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Rochester, New York, NY
- 7Division of Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, NC
Abstract
Purpose: The National Lung Screening Trial (NLST) found that screening with annual low-dose computed tomography (LDCT) reduces lung cancer (LC) mortality. Because studies have shown that the US population eligible for screening differs from NLST, we sought to quantitatively extend the NLST findings to this population.
Methods: We reweighted individual-level NLST data using inverse odds weights (IOW) to the covariate distribution of a contemporary screening-eligible population from the 2022 Behavioral Risk Factor Surveillance System (BRFSS) data. In the IOW and unweighted NLST population, we estimated the LC-specific and all-cause mortality among individuals undergoing three annual screenings of LDCT compared with chest radiography, as well as false-positive rates in the LDCT arm only.
Results: We included 53,454 NLST participants and 7,265,314 survey-weighted screening-eligible BRFSS respondents (unweighted sample of 15,779). The BRFSS population was older, more racially and ethnically diverse, and more likely to currently smoke and have respiratory comorbidities. We estimated similar benefits of screening on LC and all-cause mortality in the weighted versus unweighted NLST population (5-year LC risk difference [RD; 95% CI], –29 [–54, 4] v –23 [–44, 3]; all-cause RD [95% CI], –44 [–95, 8] v –37 [–66, –7] deaths per 10,000 screened). Estimated false-positive rates were higher in the weighted versus unweighted NLST population across three annual screenings.
Conclusion: We estimated similar mortality benefits and slightly increased false-positive rates from LDCT in the weighted versus unweighted NLST population. Our findings account for the potential differences within subgroups of interest but not variations in longitudinal adherence and diagnostic work-up between trial and clinical practice.
Related
- Understanding older adults’ perceptions and attitudes toward digital health and technology in a lung cancer screening-eligible population: a qualitative study
- HPV Vaccination Uptake Among Vaccine-Eligible U.S. Adults with and Without a Cancer History: A National Health and Nutrition Examination Survey Analysis, 2011-2023.
- Dual-Eligible Patient Burden Is Independently Associated With Lower Standardized Reimbursement Among Lumbar Laminectomy Surgeons: A National Medicare Analysis
- Twenty-Five Years of Colorectal Cancer Screening in the Czech Republic: National Registry Trends in Program Performance and Cancer Burden
- Interval colorectal cancer risk by haemoglobin concentration in Australia's National Bowel Cancer Screening Program
BibTeX
@article{su2026transporting,
title = {{Transporting Lung Cancer Screening Effects From National Lung Screening Trial to a US Screening-Eligible Population}},
author = {I-Hsuan Su and Louise M. Henderson and Xiaomeng Chen and Michael G. Hudgens and Daniel S. Reuland and M. Patricia Rivera and Courtney Schlusser and Til Stürmer and Michael Webster-Clark and Jennifer L. Lund},
journal = {JCO Clinical Cancer Informatics},
year = {2026},
month = oct,
doi = {10.1200/cci-25-00280},
url = {https://doi.org/10.1200/cci-25-00280},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Transporting Lung Cancer Screening Effects From National Lung Screening Trial to a US Screening-Eligible Population
AU - I-Hsuan Su
AU - Louise M. Henderson
AU - Xiaomeng Chen
AU - Michael G. Hudgens
AU - Daniel S. Reuland
AU - M. Patricia Rivera
AU - Courtney Schlusser
AU - Til Stürmer
AU - Michael Webster-Clark
AU - Jennifer L. Lund
PY - 2026
DA - 2026/10/26
JO - JCO Clinical Cancer Informatics
DO - 10.1200/cci-25-00280
UR - https://doi.org/10.1200/cci-25-00280
AB - Purpose: The National Lung Screening Trial (NLST) found that screening with annual low-dose computed tomography (LDCT) reduces lung cancer (LC) mortality. Because studies have shown that the US population eligible for screening differs from NLST, we sought to quantitatively extend the NLST findings to this population. Methods: We reweighted individual-level NLST data using inverse odds weights (IOW) to the covariate distribution of a contemporary screening-eligible population from the 2022 Behavioral Risk Factor Surveillance System (BRFSS) data. In the IOW and unweighted NLST population, we estimated the LC-specific and all-cause mortality among individuals undergoing three annual screenings of LDCT compared with chest radiography, as well as false-positive rates in the LDCT arm only. Results: We included 53,454 NLST participants and 7,265,314 survey-weighted screening-eligible BRFSS respondents (unweighted sample of 15,779). The BRFSS population was older, more racially and ethnically diverse, and more likely to currently smoke and have respiratory comorbidities. We estimated similar benefits of screening on LC and all-cause mortality in the weighted versus unweighted NLST population (5-year LC risk difference [RD; 95% CI], –29 [–54, 4] v –23 [–44, 3]; all-cause RD [95% CI], –44 [–95, 8] v –37 [–66, –7] deaths per 10,000 screened). Estimated false-positive rates were higher in the weighted versus unweighted NLST population across three annual screenings. Conclusion: We estimated similar mortality benefits and slightly increased false-positive rates from LDCT in the weighted versus unweighted NLST population. Our findings account for the potential differences within subgroups of interest but not variations in longitudinal adherence and diagnostic work-up between trial and clinical practice.
SN - 2473-4276
ER -